Viral serotypes have evolved diverse tissue tropisms through natural selection and an iterative process of genetic recombination and mutagenesis. Within this framework, the recent identification of Adeno-Associated Virus (AAV) serotypes with broad tissue tropisms has provided the gene therapy community with a versatile toolkit of AAV vectors. Our goal is to establish a thorough understanding of the structure-function correlates of the diverse tissue tropisms of AAV serotypes. To achieve such, we have devised a comprehensive, two-pronged approach to unravel structural attributes of AAV1-9, while simultaneously exploiting these serotypes as "blueprints" for novel AAV vector design. The approach exploits the ability of DNA shuffling to rapidly evolve novel phenotypes derived from parental serotypes followed by rational manipulation of novel AAV mutants to establish structural attributes at the amino acid level. The first strategy involves generation of a combinatorial AAV library through DNA shuffling of AAV serotype capsid sequences followed by directed evolution of cell type/receptor-specific mutants. The objective of this approach is to eliminate bias in the identification of so-called "hot spots" on the AAV capsid that impart a specific phenotype. The second approach is concerned with rational manipulation of such specific regions on AAV serotype capsids using site-directed mutagenesis. The goal of the latter strategy is to establish structure-function correlates of the AAV capsid at the amino acid level. The two complementary strategies are expected to generate critical structural information that will lay the groundwork for custom design of tissue-targeted AAV vectors. Research design involves the (1) synthesis of an AAV library through aforementioned strategies, directed evolution of cell type/receptor-specific AAV mutants, (2) characterization of such novel variants using molecular modeling, cryo-EM, a battery of biochemical assays in vitro, and (3) their translation into vectors for independent gene delivery applications in vivo. [unreadable] [unreadable] [unreadable]